Update app.py

app.py

@@ -17,57 +17,88 @@ model = joblib.load(model_dir + "/car_prices_model.pkl")
 
 
 
-def car(km_driven, mileage, engine, max_power, seats, age, seller_type_dealer, seller_type_individual, seller_type_trustmark_dealer,
-        fuel_type_cng, fuel_type_diesel, fuel_type_electric, fuel_type_petrol, transmission_type_automatic, transmission_type_manual):
-    input_list = []
-    input_list.append(km_driven)
-    input_list.append(mileage)
-    input_list.append(engine)
-    input_list.append(max_power)
-    input_list.append(seats)
-    input_list.append(age)
-    input_list.append(seller_type_dealer)
-    input_list.append(seller_type_individual)
-    input_list.append(seller_type_trustmark_dealer)
-    input_list.append(fuel_type_cng)
-    input_list.append(fuel_type_diesel)
-    input_list.append(fuel_type_electric)
-    input_list.append(fuel_type_petrol)
-    input_list.append(transmission_type_automatic)
-    input_list.append(transmission_type_manual)
-    # 'res' is a list of predictions returned as the label.
-    df = pd.DataFrame(input_list)
-    # df = pd.get_dummies(data=df, columns=['seller_type', 'fuel_type', 'transmission_type'])
-    # df = df.rename(columns = {'seller_type_Trustmark Dealer': 'seller_type_Trustmark_Dealer'})    
-    res = model.predict(df.T)[0]           
-    return res
-        
+def car(km_driven, mileage, engine, max_power, seats, age, seller_type, fuel_type, transmission_type):
+  input_list = []
+  input_list.append(km_driven)
+  input_list.append(mileage)
+  input_list.append(engine)
+  input_list.append(max_power)
+  input_list.append(seats)
+  input_list.append(age)
+  input_list.append(seller_type)
+  input_list.append(fuel_type)
+  input_list.append(transmission_type)
+
+  if (input_list[6] == "Dealer"):
+    input_list.pop(6)
+    input_list.insert(6, 1)
+    input_list.insert(7, 0)
+    input_list.insert(8, 0)
+  if (input_list[6] == "Individual"):
+    input_list.pop(6)
+    input_list.insert(6, 0)
+    input_list.insert(7, 1)
+    input_list.insert(8, 0)
+  if (input_list[6] == "Trustmark Dealer"):
+    input_list.pop(6)
+    input_list.insert(6, 0)
+    input_list.insert(7, 0)
+    input_list.insert(8, 1)
+
+  if (input_list[9] == "CNG"):
+    input_list.pop(9)
+    input_list.insert(9, 1)
+    input_list.insert(10, 0)
+    input_list.insert(11, 0)
+    input_list.insert(12, 0)
+  if (input_list[9] == "Diesel"):
+    input_list.pop(9)
+    input_list.insert(9, 0)
+    input_list.insert(10, 1)
+    input_list.insert(11, 0)
+    input_list.insert(12, 0)
+  if (input_list[9] == "Electric"):
+    input_list.pop(9)
+    input_list.insert(9, 0)
+    input_list.insert(10, 0)
+    input_list.insert(11, 1)
+    input_list.insert(12, 0)
+  if (input_list[9] == "Petrol"):
+    input_list.pop(9)
+    input_list.insert(9, 0)
+    input_list.insert(10, 0)
+    input_list.insert(11, 0)
+    input_list.insert(12, 1)
+
+  if (input_list[13] == "Automatic"):
+    input_list.pop(13)
+    input_list.insert(13, 1)
+    input_list.insert(14, 0)
+  if (input_list[13] == "Manual"):
+    input_list.pop(13)
+    input_list.insert(13, 0)
+    input_list.insert(14, 1)
+  df = pd.DataFrame(input_list)
+  res = model.predict(df.T)[0]
+  return res
+    
 demo = gr.Interface(
-    fn=car,
-    title="Car Price Predictive Analytics",
-    description="Experiment with car details to predict the price of your car.",
-    allow_flagging="never",
-    inputs=[
-        gr.inputs.Number(default=10000.0, label="Kilometers driven"),
-        gr.inputs.Number(default=12.0, label="Mileage (in KM/L)"),
-        gr.inputs.Number(default=1199.0, label="Engine Size (in cc)"),
-        gr.inputs.Number(default=70.0, label="Max Power (in BHP)"),
-        gr.inputs.Number(default=5, label="Number of Seats"),
-        gr.inputs.Number(default=4.0, label="Age of the car"),
-        gr.inputs.Number(default=0, label="seller_type_dealer"),
-        gr.inputs.Number(default=0, label="seller_type_individual"),
-        gr.inputs.Number(default=0, label="seller_type_trustmark_dealer"),
-        gr.inputs.Number(default=0, label="fuel_type_cng"),
-        gr.inputs.Number(default=0, label="fuel_type_diesel"),
-        gr.inputs.Number(default=0, label="fuel_type_electric"),
-        gr.inputs.Number(default=0, label="fuel_type_petrol"),
-        gr.inputs.Number(default=0, label="transmission_type_automatic"),
-        gr.inputs.Number(default=0, label="transmission_type_manual"),
-        
-        # gr.inputs.Dropdown(choices=["Dealer", "Individual", "Trademark Dealer"]),
-        # gr.inputs.Dropdown(choices=["Petrol", "Diesel", "Other"]),
-        # gr.inputs.Dropdown(choices=["Automatic", "Manual"]),
-        ],
-    outputs=gr.Number())
+  fn=car,
+  title="Car Price Predictive Analytics",
+  description="Experiment with car details to predict the price of your car.",
+  allow_flagging="never",
+  inputs=[
+      gr.inputs.Number(default=10000.0, label="Kilometers driven"),
+      gr.inputs.Number(default=12.0, label="Mileage (in KM/L)"),
+      gr.inputs.Number(default=1199.0, label="Engine Size (in cc)"),
+      gr.inputs.Number(default=70.0, label="Max Power (in BHP)"),
+      gr.inputs.Number(default=5, label="Number of Seats"),
+      gr.inputs.Number(default=4.0, label="Age of the car"),
+      
+      gr.inputs.Dropdown(choices=["Dealer", "Individual", "Trustmark Dealer"]),
+      gr.inputs.Dropdown(choices=["Petrol", "Diesel", "Electric", "CNG"]),
+      gr.inputs.Dropdown(choices=["Automatic", "Manual"]),
+      ],
+  outputs=gr.Number())
 
 demo.launch()